Secondary Logo

Evaluation of a new upper body ergometer for cross-country skiers

WISLØFF, ULRIK; HELGERUD, JAN

Medicine& Science in Sports & Exercise: August 1998 - Volume 30 - Issue 8 - p 1314-1320
Special Communications: Methods
Free

Evaluation of a new upper body ergometer for cross-country skiers. Med. Sci. Sports Exerc., Vol. 30, No 8, pp. 1314-1320, 1998. A new specific ski ergometer has been developed to study aerobic endurance and force development in the upper body of cross-country skiers. The major purpose of the present study was to examine the validity and the reliability of this ergometer. Eleven male cross-country skiers participated in the study. Work on the ski ergometer, at an inclination of 4°, incorporated the double-poling technique. All subjects participated in three tests in addition to a pretest where peak oxygen uptake in the upper body (V˙O2peak) and maximal oxygen uptake during treadmill running (V˙O2max) were measured. In a field-test, subjects performed double poling uphill, and V˙O2peak was reached after 4-6 min. There was no statistically significant differences in power output (Watt) or oxygen uptake (V˙O2) at the same exercise stages between tests and the coefficient of variation was 2.0% and 2.5%, respectively. There were no significant differences between V˙O2peak as measured in the field and on the ski ergometer. The average upper body/leg ratio was 90% (range 78.1-97.1%). There was no significant correlation between V˙O2max and V˙O2peak. The present study showed the ski ergometer to be both reliable and valid for evaluating V˙O2 and force development in the upper body at submaximal and maximal workloads for cross-country skiers.

Department of Sport Sciences, Norwegian University of Science and Technology, N-7055 Dragvoll, Trondheim, NORWAY

Submitted for publication May 1996.

Accepted for publication September 1996.

Address for correspondence: Jan Helgerud, Department of Sport Sciences, Norwegian University of Science and Technology, N-7055 Dragvoll, Trondheim, Norway. E-mail: Janhe@sv.ntnu.no.

Cross-country skiing is characterized by repeated contraction of a number of muscles over extended periods of time. In the last decades techniques have changed so that upper body strength and aerobic endurance seem to play a still more important role. This has lead to more training and testing of the upper body (14,17,23). Peak oxygen uptake, measured on small muscle groups, can change considerably with training without any commensurate changes in V˙O2max as measured while running on a treadmill (5,9). For this reason, there is need for both a high V˙O2max and high a V˙O2peak of the muscles involved in the various skiing techniques. Bergh (4) and Ingjer (10) have recommended the use of V˙O2max in ml·kg−0.67·min−1 to indicate the performance capacity of cross-country skiers because the oxygen cost involved does not increase proportionally to body mass; i.e., the oxygen uptake per kg body mass decreases as body mass increases. Bergh (3) has indicated that cross-country skiers seldom use more than 10-20% of their maximal strength in any single contraction during classical technique. However, there are situations where the force requirements are considerably greater. In addition, the greater poling force used in the skating technique compared with the classical technique(22,25) suggests that upper body strength is more important when the skating technique is used. This is supported by production of higher blood lactate ([la]b) during skating as compared with the classical technique (19). Furthermore, it has been suggested that increased muscle strength may increase endurance performance (7,14,26). It is not surprising, therefore, that quadriceps and upper body strength have been found to significantly correlate with skiers performances in cross-country races (20).

The arm crank ergometer is the most commonly used type for testing aerobic capacity in the upper body. The criticism of this ergometer, when testing cross-country skiers, is that it has a different movement pattern and does not include all relevant muscle groups used in cross-country skiing. There exist additional types of ergometers for simulation of cross-country skiing, but few of them seem specific enough to evaluate upper body aerobic capacity (18).

Mygind et al. (18) measured V˙O2max during treadmill running and V˙O2peak during double poling on a ski ergometer to evaluate which test correlated best with performance in cross-country skiing. There was no significant correlation between V˙O2max and V˙O2peak and only double poling correlated significant with performance. The upper body/leg ratio increased by 8% in 3 months, but V˙O2max did not change significantly during the same period. They also found marked differences in V˙O2peak for skiers with approximately the same V˙O2max, a finding that indicates that important parameters are missed when only V˙O2max is tested. These results emphasize the importance of a standardized ergometer and protocol for testing aerobic capacity and force development in the upper body of cross-country skiers.

A drawback with Mygind et al.'s (18) ergometer is that the legs are fixed at a flat base, thereby losing the natural phase of acceleration and movement of the body as in cross-country skiing. For this reason, an alternative ergometer that takes these factors into account has been constructed. The major purpose of the present study was to examine the validity and reliability of this new, specific, ski ergometer for evaluating aerobic endurance and force development in the upper body of cross-country skiers.

Back to Top | Article Outline

MATERIALS AND METHODS

Eleven male cross-country skiers participated in the study. Their mean age was 21 (3.8) yr, height 178 cm (6.6), weight 72 kg (2.4), maximal heart rate (fcmax) 195 (5.9), and V˙O2max 70.1 ml·kg−1·min−1(2.8) (mean ± SD). Each subject reviewed and signed consent forms approved by the Human Research Review Committee prior to participating.

Ski ergometer. A ski ergometer to evaluate aerobic endurance in the upper body of cross-country skiers was developed by Helgerud and Rasmussen (Fig. 1). It consists of a platform on which the subject can stand and two poles, with wheels running on rails, connected to an electric motor (Elektri 0.75 KW, Sg 80-4B, Norway) in front of the ergometer by wires. Above the motor, there was a system for elastic wires which turned on the motor drum when the poles were pushed backwards and thereby facilitating the forward movement of the poles. Assisting this movement was necessary because the ball-bearings of the poles made them heavier than normal ones. The wires were very long, so that the extra resistance induced by this system was approximately the same, independent of pole position. The ergometer allows use of the double poling technique as well as individual arm action as in classic technique. The ergometer had an inclination of 4° and the resistance was changed by changing the speed of the motor that, in turn, was controlled by a computer. The power output expressed in Watts (W) was calculated from the equation: Equation[1] where Mb is the body mass in Newtons, ν is the motor speed in m·s−1, and α is the inclination of the ski ergometer.

Figure 1

Figure 1

The poles were braked by the motor when poling speed exceeded the motor speed, and, thereby, power was developed and the platform was accelerated. Power output was registered by a load cell (Tefka 250 kg, 1.8 mV/V, Denmark) placed in front of the ergometer, and, therefore, registered all power output, friction of the system included. The load cell was connected to the computer on which data were recorded and was linear from 0 to 250 kg with a reproducibility of 0.1% (12). The load cell was calibrated using a dynamometer (Dynamometer NO 22, Germany) with 0, 10, and 20 kg. The dynamometer had an accuracy of 0.1%. The motor speed was calibrated using a tachometer with an accuracy of 0.1%. The subjects were able to move their legs on the platform in such a way that the upper body movement would be similar to the actual movements performed when double poling in cross-country skiing.

Procedure. Each subject performed a preliminary test on the ski ergometer and on the treadmill (Jaeger LE-5000) in order that V˙O2peak and V˙O2max, respectively, could be measured. The subjects used poles of the same length as they would for the classical technique. All subjects participated in three tests on the ski ergometer in addition to the pretest. Six of the subjects also performed a field test of cross-country skiing. All tests were carried out during a period of 14 d.

The subjects worked at exercise stages, 50, 60, and 70% of V˙O2peak (referred to as stages 1, 2, and 3) each time for 4 min, with 30-s pause between each trial to take a blood sample before the power was increased 20 W every minute to reach V˙O2peak. The blood lactate concentration ([la]b), unhemolyzed, was measured by a YSI Model 1500 Sport Lactate Analyzer (Yellow Springs Instrument Co., Yellow Springs, OH). In test 1, five subjects received the exercise stages in increasing order (stages 1, 2, 3), and six got decreasing order (stages 3, 2, 1). In test 2, the stages were given in the reversed manner while in test 3 the orders were the same as in test 1. Each subject had a warm-up at 40-50% of V˙O2peak for 15 min, then the exercise stages were given in increasing or decreasing order. At each level, the power output was recorded by the load cell. Heart rate (fc) was measured by short range radio telemetry (Polar Sporttester, Polar Electro, Finland). Oxygen uptake (V˙O2), minute ventilation (V˙E), and breathing frequency (fb) were measured during each exercise stage using an Ergo Oxyscreen (Jaeger EOS Sprint, Germany) and the measurements between 2.5 and 3.5 min were recorded. In the field-test, subjects performed double poling going uphill (approximately 4° inclination) in a well-prepared track, so that V˙O2peak was reached after 4-6 min. The ambient temperature was −5°C during the field-test. For measuring V˙O2 in the field-test, Cosmed K2 (Cosmed K2, Italy) was used. The K2 system consisted of a mouthpiece with a turbin flowmeter, a transmitter, a battery, and a receiver. In the flow meter, there was a photo-detector that measured the velocity of revolution of the turbine, and from the velocity, flow volume was calculated. The K2 had a receiver to pick up fc readings from the Polar Sporttester. V˙O2 was calculated at the receiver from the following equation:Equation [2]E at body temperature and pressure, saturated with water vapor (BTPS) was determined at the transmitter. Part of the expired air was sampled from a pick-up located near the turbine and fed into the transmitter, where oxygen concentration of expired air (FEO2) was detected by a polarographic oxygen sensor. Total mass of the K2 that the subject carried was about 850 g. The telemetry transmission range was about 100 m, and subjects were followed by snowscooter during the test. With the K2, the calculation of V˙O2 was made under the assumption that respiratory exchange ratio (R) was 1. When R is lower than 1, V˙O2 is underestimated, and when it is higher than 1, it is overestimated. At V˙O2peak in the field-test, there is good reason to believe that R was higher than 1, and to correct for this the following formula was used (21): Equation [3] The R-values used in the calculation were V˙O2peak values from test 3 at the ski ergometer.

Statistical procedures. The results are shown as means (X¯), ranges, standard deviation (SD), and coefficient of variation (V) calculated with conventional procedures. Pearson correlation coefficients, paired t-tests, and ANOVA repeated measurement analysis were used to determine differences between tests at each exercise stage. Results were accepted as significant at P < 0.05.

Back to Top | Article Outline

RESULTS

There were no statistically significant differences in power output (W) performed at the same exercise stage between tests. Average test-retest results from test 2 and 3 are presented in Table 1.

TABLE 1

TABLE 1

Figure 2 illustrates the force parameters displayed on the screen during tests on the ski ergometer. Each curve represents a double poling. Peak force and time to peak force were recorded and mean power calculated at each exercise stage.

Figure 2

Figure 2

No statistically significant differences were found for V˙O2 at the same power output between tests, but no significant correlation for V˙O2 at submaximal exercise stages between tests 1 and 3 was found. There was high correlation for V˙O2 at the same exercise stages between tests 2 and 3 (Fig. 3). The coefficient of variation was around 2.5% at every exercise stage. Average V˙O2peak from the preliminary test was 4.42 L·min−1 (0.74) and not significantly different from V˙O2peak on any other test.

Figure 3

Figure 3

There was a linear increase in V˙O2 (L·min−1) with increased power output (W) (r = 0.93, P < 0.001) (Fig. 4).1/2pick;f4;0;1pick;1852f4;0;;;ZPICKFOOT;F4 When the relation between V˙O2 and power output was calculated for each person the average r become 0.997 (0.002) (P < 0.001).

Figure 4

Figure 4

V˙O2max was significantly higher than V˙O2peak(Table 2) expressed as L·min−1 and in relation to body mass (mL·kg−1·min−1 or mL·kg−0.67·min−1). The average upper body/leg ratio was 90% (range 78.1-97.1%). There was no significant correlation between V˙O2max and V˙O2peak, and there were marked differences in V˙O2peak for skiers with approximately the same V˙O2max. This is illustrated by two skiers having a V˙O2max of 4.8 L·min−1 with upper body/leg ratios of 78 and 90%, respectively. No significant differences were found for V˙E and [la]b at V˙O2max and V˙O2peak for the entire group. There were no significant differences between V˙O2peak as measured in the field and on the ski ergometer (Table 3), while peak minute ventilation (V˙Epeak) and peak heart rate (fcpeak) were significantly higher at V˙O2peak on the ski ergometer compared to the field test. Averaged V˙O2peak in the field test corrected for R was 3.9 L·min−1 (0.33) and was not significantly different from any test on the ski ergometer.

TABLE 2

TABLE 2

TABLE 3

TABLE 3

Back to Top | Article Outline

DISCUSSION

Validity. The present study showed the ski ergometer to be valid for evaluating V˙O2 at submaximal and maximal exercise stages in the upper body of cross-country skiers. After the preliminary test, three tests were performed because of a possible learning effect following on test 1, caused by unfamiliarity with the apparatus and possible discomfort wearing the test equipment during poling. This was not necessary when measuring V˙O2peak since there were no significant differences between V˙O2peak on the preliminary test and any of the other tests. A methodological problem was introduced by using Jaeger EOS-sprint to measure V˙O2 on the ski ergometer and Cosmed K2 in the field-test, since Cosmed K2 sets R to 1 and overestimates V˙O2peak if R is in reality higher. Cosmed K2 has been found, by others, to be a reliable and valid apparatus for measurement of oxygen uptake (6,13). To correct V˙O2 measured with the Cosmed K2, equation 1 was used with R-values from test 3 on the ski ergometer. However, we still do not know whether these values were representative for the real R in the field-test, but it is reasonable to believe that the R-values were close to 1.10 as on the ski ergometer. V˙O2peak measured on the ski ergometer (N = 6) was 4.08 L·min−1 (0.24) and was not significantly different from V˙O2peak in the field-test (N = 6) which was 3.91 L·min−1(0.35). The ski ergometer is considered to be very spesific for cross-country skiing as indicated by the perception of the skier subjects.

Reliability. The present study showed the ski ergometer to be reliable for evaluating V˙O2 in the upper body of cross-country skiers. Test-retest correlation for V˙O2 was high (r = 0.98, P < 0.001). The most obvious test for evaluating the ski ergometer was to perform a double poling test in the field, while the subjects were cross-country skiing. The correlation between the two tests was high (r = 0.93,P < 0.001). This alternative-form reliability provides a useful measure for evaluating tests (1).

From a metabolical point of view V˙O2 was expected to be the same at corresponding exercise stages in different tests, regardless of whether the exercise stages were given in an increasing or decreasing manner. The reason for choosing 4 min at each exercise stage was that it normally takes 2-3 min to reach a steady rate V˙O2 after changing the power output (2). The risk of starting at 70% of V˙O2peak was that it could be over the anaerobic threshold in the upper body of the test subjects. If this were to be the case, differences in V˙O2 between increasing and decreasing exercise stages could be expected. V˙O2 at a lower exercise stage could then be higher because an oxygen deficit had been developed, causing a higher cost of poling (2). For all subjects in the present study one had reason to believe that 70% of V˙O2peak was lower than anaerobic threshold in upper body compared with corresponding values in running (7). A continuous protocol was used because others (28) did not find differences in V˙O2peak between using discontinuous or continuous protocols. A pilot study had shown that an increase of 20 W on the ski ergometer gave a change in V˙O2 of approximately 5 mL·kg−1·min−1, which also corresponded with the increase of 15 W as obtained by Mygind et al. (18) in a continuous protocol for measuring V˙O2peak. A plateau for V˙O2 was observed for all subjects, while testing V˙O2peak, despite of further increase in the rate of exercise during work on the ski ergometer and in the field-test, and is therefore a natural criterion for reaching V˙O2peak. When measuring V˙O2peak in the field, care was taken to choose moderately steep hills so that use of maximal force would not be necessary. Use of maximal force would squeeze muscle capillaries, which, in turn, would obstruct local blood flow and limit peripheral oxygen transport (15,24).

If observed power output varied much for the same exercise stages in different tests, the ergometer would not have been of much use. Correlation between power output at the same exercise stages in tests 2 and 3 was high (r = 0.995). Theoretically, the same power output were expected from one test to another when the same values were used in equation 1. Small differences in observed watt between tests were observed and may have been caused by friction between the wheels of the platform and the rail, which may differ somewhat depending on the exact location of the subject on the platform. By standing at the front or back of the platform, the friction may increase a little because of more intense pressure on the wheels and increase the power output somewhat. The drawback of increased friction, when the platform is rolling forward, during the double poling is compensated for by slower rolling of the platform backward during preparation for the next poling. The same is true when the subject stands in the middle of the platform. To avoid variation between tests, the exact location of subjects on the platform could have been controlled by instructing subjects where to position themselves. This was not done because one wanted the subjects to move their legs as naturally as when skiing. The tightness of the elastic wires connected to the poles may have changed during the test period and lead to changes in observed power output between tests. In the same manner, the friction between the pole wheels and the rail could change, depending on how well the rail was greased, and lead to small changes in observed power output. The 2.0% coefficient of variation must be considered as small and did not lead to significant changes in V˙O2 at any exercise in the present study.

No significant differences for V˙O2 at the same exercise stages between any test was as expected, since there were no significant differences between observed power output(W) at the same exercise stages between tests. No significant correlation at submaximal exercise stages between tests 1 and 3 indicates the need for a preliminary test when the aim is to compare V˙O2 at submaximal power outputs. Without a pretest, the reason for changes in V˙O2 at the same exercise stages from one test to another could be the result of learning. No significant differences in V˙O2 at the same exercise stages and the high correlation between tests 2 and 3 mean that it is possible to compare results from one test with another. There was a 2.5% coefficient of variation in VO2 at the same exercise stage, and it was within the same magnitude as the reliability of Ergo Oxyscreen (11, 27) when measuring V˙O2(11). The linear increase in V˙O2, with increasing power output (average r = 0.997 (0.002), P< 0.001) is normal for continuous dynamical work of large muscle groups (2,16).

V˙O2peakversus V˙O2max. V˙O2peak on the ski ergometer was approximately 90% of V˙O2max when running on the treadmill and was almost the same as Mygind et al. (18) found in their study. This is higher than others have found during arm cranking or classic arm action but is reasonable considering the extensive use of trunk muscles in double poling. In addition, the subjects in these two studies are better trained, considering V˙O2max and V˙O2peak, than most other studies. In the present study with an average V˙O2peak(N = 11) of 4.55 L·min−1 (0.76), the power output was approximately 180 W. To achieve the same V˙O2 (4.90 L·min−1(0.30)) on Mygind et al's. (18) ski ergometer, the power output had to be 215 W. This could indicate use of a greater muscle mass in that study. On the other hand, the subjects could have had more upper body strength and therefore tolerated greater power output at V˙O2peak. If this was the case, the subjects would perhaps have reached V˙O2peak at 180 W if enough time was given at that exercise stage.

The present study showed no significant correlation between V˙O2max and V˙O2peak, and there were marked differences in V˙O2peak inspite of approximately the same V˙O2max. This agrees with the results of Mygind et al. (18). These are important results that do not show up during the traditional test of V˙O2max and that emphasize the importance of testing aerobic endurance in the upper body of cross-country skiers. The argument becomes even stronger considering that Mygind et al. (17) did not find a significant correlation between V˙O2max and performance but between V˙O2peak and performance. Combined with results from studies which indicate that increased muscle strength may increase aerobic endurance (8,14,26), results from the present study and the study of Mygind et al. (18) indicate that more attention should be paid to the training and testing of upper body strength and endurance of cross-country skiers.

There were no significant differences for [la]b at V˙O2max and V˙O2peak despite significant differences in V˙O2 and power output between these two tests. [la]b may be dependent on total work done (2,16) and the central apparati for oxygen transport, and, therefore, it would be reasonable to expect higher[la]b at the V˙O2max-test, which incorporated the greatest muscle mass of these two tests. [la]b is also affected by the exercising muscles ability to utilize the oxygen delivered. There is good reason to believe that, for a well-trained individual, V˙O2peak is limited by the small muscle mass involved, restricted capillary density, the mean transit time of muscle blood flow, and smaller oxidative capacity (24). A small muscle mass may lead to use of maximal force and result in high intramuscular pressures that would limit perfusion of the working muscle. A reduction in blood flow to working muscles led to an increase in anaerobic metabolism and lactate production(15,24). Considering the metabolic characteristics above, a [la]b over 6-8 mmol·L−1 could be used as a criterion for reaching V˙O2peak on the ski ergometer as it is for reaching V˙O2max(2). No significant differences in [la]b between the field-test and work on the ski ergometer indicate that the amount of anaerobic work in these two tests was approximately the same. On the ski ergometer fcpeak (N = 6) was 186(9.1) beats·min−1 and significantly higher than fcpeak in the field-test, which was 178 (8.6) beats·min−1. This difference may be explained by the fact that subjects worked under quite different ambient conditions in these two tests. The ambient temperature during work on the ski ergometer was 20°C while the temperature was −5°C during the field test. Another explanation may be more dynamical work with legs in the field test and, thereby, better filling of the heart and increased stroke volume compared with work on the ski ergometer.

No significant differences were found between V˙Epeak and V˙Emax on the ski ergometer and treadmill, respectively. This may have been brought about by the fact that it is not necessary to reach V˙Emax in order to reach V˙O2max(28). On the other hand, the power output on the ski ergometer could be high enough to reach V˙Emax. V˙Epeak on the ski ergometer (N = 6) was 153.7 (4.9) L·min−1 and significantly higher than V˙Epeak in the field-test, which was 133.3(6.2) L·min−1. This was not expected since no significant differences in either VO2, [la]b, or breathing frequency were found between the tests. Finding a lower V˙Epeak in the field-test probably reflects that V˙O2peak can be reached without reaching V˙Epeak. Another explanation may be that the K2 system underestimates values for V˙E. If this was the case the values for V˙O2 also must be too low since V˙E was used for calculating V˙O2. Others (13) have found the ventilation system of the K2 to have a good reproducibility for values up to 180 L·min−1.

Conclusion. The present study showed the ski ergometer to be valid and reliable for evaluating V˙O2 in the upper body at submaximal and maximal exercise stages of cross-country skiers. No significant differences for power output (W) or V˙O2 (L·min−1), at the same exercise stages, between tests were observed. Coefficient of variation of 2.0% and 2.5% for power output and V˙O2, respectively, were observed. There were no significant differences between V˙O2peak in the field test and on the ski ergometer. The absence of significant correlation between V˙O2max and V˙O2peak highlights the need for a specific test of the upper body of cross-country skiers.

Practical applications. The ergometer used for the present study has been modified so that it is possible to change the inclination by a computer. The ski ergometer is versatile in that it allows individual arm action and is adjustable for leg actions as in classical technique in cross-country skiing. This give the opportunity for one-leg or one-arm studies for evaluating specific training effect. For evaluating strength parameters involved in the movement of the arms, both in double poling and classical technique, force parameters as peak force, time to peak force, average power and cycle length, and rate of the stride may be registered. The ergometer is also adjustable to testing or training of groups, other than cross-country skiers, where strength and endurance in the upper body are important factors. It is easy to adjust the ergometer for spesific testing of rowers, kayakers, swimmers, wheel chair users, and perhaps other groups as well.

Trying out test protocols with different inclinations will be done, as well as developing a protocol for testing anaerobic threshold in the upper body. Studies of training endurance and maximal strength in the upper body to investigate changes in aerobic endurance, heart function, and dimensions at maximal and submaximal exercise stages are already being performed.

Back to Top | Article Outline

REFERENCES

1. Anastasia, A. Psychological Testing, 6th Ed. New York: Macmillan, 1988, pp. 115-116.
2. Åstrand, P.-O., and K. Rodahl. Textbook of Work Physiology. New York: McGraw-Hill, 1986.
3. Bergh, U. Physiology of Cross-Country Ski Racing. Champaign, IL: Human Kinetics, 1982.
4. Bergh, U. The influence of bodymass in cross-country skiing. Med. Sci. Sports Exerc. 19:324-331, 1987.
5.Clausen, J. P. Circulatory adjustments to dynamic exercise and effect of physical training in normal subjects and in patients with coronary disease. Prog. Card. Dis. 18:459-495, 1976.
6. Dal Monte, A. Maximum oxygen consumption by telemetry. Sports Culture Rev. 3-12, 1989.
7. Helgerud, J. Maximal oxygenuptake, anaerobic threshold and running performance in women and men with similar performances levels in marathons. Eur. J. Appl. Physiol. 68:155-161, 1994.
8. Hixon, R. D., M. A. Rosenkoetter, and M. M. Brown. Strength training effects on aerobic power and short term endurance. Med. Sci. Sports Exerc. 12:336-339, 1980.
9. Holmer, I. Physiology of swimming man. Acta Physiol. Scand. Suppl. 407, 1974.
10. Ingjer, F. Maximal oxygen uptake as a predictor of performance ability in woman and man elite cross-country skiers. Scand. Med. Sport Exerc. 1:25-30, 1991.
11.Instruction Manual. EOS sprint version 3.0. First ed. GmbH and CoKF, Würsburg, Germany, 1988.
12. Instruction Manual. Scan-Sense, Husøysund, Norway, 1988.
13. Kawakami, Y., D. Nozaki, A. Matsuo, and T. Fukunaga. Relability of measurement of oxygen uptake by a portable telemetric system. Eur. J. Appl. Physiol. 65:409-414, 1992.
14. Kelly, J. M. Physiology of cross-country skiing. In: Winter Sports Medicine, M. J. Casey, C. Foster and E. G. Hixon (Eds.). 4:227-283, 1990.
15. LeJemtel, T. H., C. S. Maskin, D. Lucido, and B. J. Chadiwicj. Failure to augment maximal limb blood flow in response to one-leg versus two-leg exercise in patients with severe heart failure. Circulation 74:245-251, 1986.
16. McArdle, W. D., F. I. Katch, and V. L. Katch. Essentials of Exercise Physiology. Philadelphia: Lea & Febiger, 1994.
17. Millerhagen, J. O., J. M. Kelly, and T. J. Murphy. A study of combined arm and leg exercise with application to nordic skiing. Can. J. Appl. Sport Sci. 8:92-97, 1983.
18. Mygind, E., B. Larsson, and T. Klausen. Evaluation of a specific test in cross-country skiing.J. Sports Sci. 9:249-257, 1991.
19. Mygind, E., L. B. Andersen, and B. Rasmussen. Blood lactate and respiratory variables in elite cross-country skiing at racing speeds. Scand. Med. Sci. Sports 4:243-251, 1994.
20. Ng, A. V., R. B. Demment, D. R. Basset, et al. Characteristics and performance of male citizen cross-country ski racers. Int. J. Sports Med. 9:205-209, 1988.
21.Operator Manual. Cosmed K2 System Vacumetrics Inc. Vacumed Division, Ventura, CA, 1992.
22. Pierce, J. P., M. H. Pope, P. Renstrøm, R. J. Johnson, J. Dufek, and C. Dillman. Force measurement in cross-country skiing.Int. J. Sports Biometr. 3:382-387, 1987.
23.Sharkey, B. J. Training for Cross-Country Ski Racing: A Physiological Guide for Athletes and Coaches. Champaign, IL: Human Kinetics Publishers, 1984.
24. Shephard, R. J., E. Bouhlel, H. Vanderwalle, and H. Monod. Muscle mass as a factor limiting physical work. J. Appl. Physiol. 64:1472-1479, 1988.
25. Smith, G. A. Kinetic analysis of the V1 skate in cross country skiing. Proceedings of the first IOC World Congress on Sport Sciences, 1989, pp. 281-282.
26. Stone, W. J. and S. P. Coulter. Strength/endurance effects from three resistance training protocols with women. J. Strength Cond. Res. 8:231-234, 1994.
27. Versteig, P. G. A. and Kippersluis, G. J. Automated systems for measurement of oxygen uptake during exercise testing. Int. J. Sports Med. 10:107-112, 1989.
28. Washburn, R. A. and D. R. Seals. Comparison of continuous and discontinuous protocols for the determination of peak oxygen uptake in armcrancing.J. Appl. Physiol. 51:3-6, 1983.
Keywords:

CROSS-COUNTRY SKIING; SKI ERGOMETER; UPPER BODY; V˙O2max; V˙O2peak

© Williams & Wilkins 1998. All Rights Reserved.